Privacy-Preserving Neural Processes for Probabilistic User Modeling

Amir Sonee, HARIPRIYA HARIKUMAR, Alex Hämäläinen, Lukas Prediger, Samuel Kaski

Keywords: differential privacy, surrogate user models, neural process, meta-learning

TL;DR: We provide theoretical and empirical guarantees for privacy-preserving probabilistic user modeling using user-level differentially private meta-learning with Neural Processes, ensuring strong privacy in security-critical AI applications.

Abstract: Uncertainty-aware user modeling is crucial for designing AI systems that adapt to users in real-time while addressing privacy concerns. This paper proposes a novel framework for privacy-preserving probabilistic user modeling that integrates uncertainty quantification and differential privacy (DP). Building on neural processes (NPs), a scalable latent variable probabilistic model, we enable meta-learning for user behaviour prediction under privacy constraints. By employing differentially private stochastic gradient descent (DP-SGD), our method achieves rigorous privacy guarantees while preserving predictive accuracy. Unlike prior work, which primarily addresses privacy-preserving learning for convex or smooth functions, we establish theoretical guarantees for non-convex objectives, focusing on the utility-privacy trade-offs inherent in uncertainty-aware models. Through extensive experiments, we demonstrate that our approach achieves competitive accuracy under stringent privacy budgets. Our results showcase the potential of privacy-preserving probabilistic user models to enable trustworthy AI systems in real-world interactive applications.

Code Link: https://github.com/AI-Fundamentals/DiffPrivNPUserModeling

Readers: auai.org/UAI/2025/Conference, auai.org/UAI/2025/Conference/Area_Chairs, auai.org/UAI/2025/Conference/Reviewers, auai.org/UAI/2025/Conference/Submission507/Authors, auai.org/UAI/2025/Conference/Submission507/Reproducibility_Reviewers

Submission Number: 507